Recent epidemiologic data shown that provision of vitamin A supplements to children at risk of developing vitamin A deficiency decreases mortality rates from infectious disease. Clinical trials have also recently shown that provision of high-dose vitamin A supplements to children with measles decreases mortality. Both field and clinical studies suggest that vitamin A supplementation specifically affects recovery from diarrheal disease. Although vitamin A has well-documented effects on the serum antibody response and on epithelial cell differentiation in the small intestine, the mechanism by which vitamin A protects against death or development of severe disease is not know. Elucidating these mechanisms will have clinical and public health implications. Population-based supplementation programs are effective but costly. It is possible that supplementation of children attending treatment centers with acute infections can improve the clinical course of disease and avert the same number of deaths that would be prevented by a community-wide program. This approach has not been attempted. The goal of this study is to examine the mechanisms of action of vitamin A supplementation in children with acute diarrhea and to determine if there is a clinically demonstrable improvement. Specific Aims are as follows: In a double-blind, placebo- controlled clinical trial conducted in Lima, Peru among pediatric inpatients with rotavirus diarrhea studies will determine if vitamin A supplementation affects (1) the immune response to rotavirus infection by measuring mucosal (salivary and fecal IgA and IgM) and serum (IgA, IgG, IgM) concentrations of rotavirus- specific antibody 1, 3, 10 and 20 days after admission; (2) the course of rotavirus disease by measuring duration of rotavirus shedding in the stool, duration of diarrhea (considering stool consistency, number and weight), and severity of diarrhea (as measured by stool weight); (3) the regeneration of normal intestinal epithelial function during recovery from rotavirus infection by measuring the prevalence of lactase insufficiency using the breath hydrogen test (BHT) 2, 10 and 20 days after admission. In addition, the vitamin A status of placebo and control subjects after recovery from infection (10 d after admission) using the relative dose-response (RDR) test to determine the prevalence of vitamin A deficiency in this clinical population will be measured.